Potential role of cardiac calsequestrin in the lethal arrhythmic effects of cocaine

Abstract

Background: Cocaine-related deaths are continuously rising and its overdose is often associated with lethal cardiotoxic effects.

Methods and results: Our approach, employing isothermal titration calorimetry (ITC) and light scattering in parallel, has confirmed the significant affinity of human cardiac calsequestrin (CASQ2) for cocaine. Calsequestrin (CASQ) is a major Ca(2+)-storage protein within the sarcoplasmic reticulum (SR) of both cardiac and skeletal muscles. CASQ acts as a Ca(2+) buffer and Ca(2+)-channel regulator through its unique Ca(2+)-dependent oligomerization. Equilibrium dialysis and atomic absorption spectroscopy experiments illustrated the perturbational effect of cocaine on CASQ2 polymerization, resulting in substantial reduction of its Ca(2+)-binding capacity. We also confirmed the accumulation of cocaine in rat heart tissue and the substantial effects cocaine has on cultured C2C12 cells. The same experiments were performed with methamphetamine as a control, which displayed neither affinity for CASQ2 nor any significant effects on its function. Since cocaine did not have any direct effect on the Ca(2+)-release channel judging from our single channel recordings, these studies provide new insights into how cocaine may interfere with the normal E-C coupling mechanism with lethal arrhythmogenic consequences.

Conclusion: We propose that cocaine accumulates in SR through its affinity for CASQ2 and affects both SR Ca(2+) storage and release by altering the normal CASQ2 Ca(2+)-dependent polymerization. By this mechanism, cocaine use could produce serious cardiac problems, especially in people who have genetically-impaired CASQ2, defects in other E-C coupling components, or compromised cocaine metabolism and clearance.

Keywords: CASQ; CPVT; Calsequestrin; Cocaine; ITC; K(d); Methamphetamine; RYR; Ryanodine receptor; SR; Sarcoplasmic reticulum; calsequestrin; catecholamine-induced polymorphic ventricular tachycardia; dissociation constant; isothermal titration calorimetry; ryanodine receptor; sarcoplasmic reticulum.

Fig. 1. Measurement of the heat released upon cocaine and methamphetamine injection by ITC

The trend of heat released by serial injections of ether cocaine (●) or methamphetamine (■). 50 μM CASQ2 was titrated with a 2 mM stock solution of cocaine or methamphetamine dissolved in ITC buffer (300 mM KCl, 10 mM MOPS, pH 7.5). The X-axis represents the molar ration of [Ligand] / [Protein]. The heat of binding was measured three times and averaged. The corresponding ΔH, ΔS and K_d values for cocain-binding are indicated.

Fig. 2. CASQ2 oligomeric states with and without cocaine

Elution profiles were monitored by multi-angle laser light-scattering (left Y-axis) and 280 nm UV absorption (right Y-axis) versus elution volume (X-axis). The solid line indicates the UV absorption profile in the absence of CaCl₂, and the dotted line represents the UV absorption profile in the presence of 1 mM CaCl₂. A) The monomer to dimer transition of CASQ2 resulting from no Ca²⁺ (solid line) to 1 mM Ca²⁺ (dotted line) in absence of cocaine. B) The transitional pattern from no Ca²⁺ (solid line) to 1 mM Ca²⁺ (dotted line) of CASQ2 in the presence of 350 μM cocaine. The molecular weights of elutes as determined by multi-angle laser light scattering are shown as dots at the center of each peak.

Fig. 3. CASQ2 Ca²⁺-binding capacity inhibition by cocaine

The number of Ca²⁺ ions bound to CASQ2 was determined through equilibrium dialysis and atomic absorption spectroscopy. Fractional occupancy (y = [bound Ca²⁺]/[total protein]) is plotted against [unbound Ca²⁺] for CASQ2 without drugs (open triangle), in the presence of 350 μM cocaine (filled circle), and in the presence of 350 μM methamphetamine (filled rectangle).

Fig. 4. Quantification of cocaine isolated from heart tissue

Relative quantification μg/mg of tissue for animals treated daily with 20 μg/g body weight of cocaine or saline treated samples. Cocaine concentrations were determined using a base-extraction procedure (pH 11.5). This basic condition, known to cause hydrolysis, was not controlled for by inclusion of an internal standard subject to similar conditions; therefore, actual tissue cocaine concentrations may be slightly higher than reported here.

Fig. 5. Surface representation of the CASQ2-cocaine complex

A) All the cocaine-binding positions as determined by docking are represented as brown spheres. B) Ribbon diagram representing the three major cocaine-binding sites. The orange ball and stick represents S1 binding, green represents the S2 binding site, and purple represents S3 binding. C) A close up of the proposed S1 binding with residues and electrostatically-interacting residues shown in orange. D) Representation of S2 binding and the local binding environment. E) Representation of the S3 binding site with its constituent residues shown. These figures were generated using Open-Source software PyMOL^™ (v1.4).

Fig. 6. Cocaine has no detectable action on RyR2 function

Endogenous CASQ was removed from the single RyR2 channels and no exogenous CASQ was added. A) Sample single channel recordings. Opening events are shown as upward deflections from the marked zero current level (left margin). Membrane potential was 0 mV. Cocaine was added to cytosolic solution. B) All-points histograms from 4-minute recordings in the presence and absence of 500 μM cocaine. C) Sample open and closed dwell time histograms in the absence or presence of 500 μM cocaine.

Fig. 7

A) Traces of observed F/F_o for C2C12 Myocytes incubated with DMSO vehicle (control), 175 μM methamphetamine, and 175 μM cocaine. B) Histogram representation showing the average Ca²⁺ waves for each dataset.